Patrick Rock 02 August, 2016
Synaptotagmin is a protein involved with the fusion of synaptic vessicles with the outer membrane. Mutations in a highly conserved sequence within the C2A and C2B domains are fatal in drosophila unless the calcium concentration is dramatically increased. To investigate the mechanism by which these mutations disrupt the function of synaptotagmin, we ran accelerated molecular dynamics simulations on the isolated C2A and C2B domains. We assesed the stability of the Calcium binding pocket formed by loops 1 and 3 by measuring the volume of the pocket, the secondary structure of the loops, and principle component analysis.
Synaptotagmin is a Ca2+ sensing protein, essential for life, involved in the exocytosis of synaptic vesicles through interaction with the SNARE (soluble N-ethylmaleimide-sensitive factor attachment protein receptor) complex. Synaptotagmin binds calcium in a cup like formation and subsequently embeds itself in the target membrane through interactions with hydrophobic residues on loops 1 and 3. Fatal AD3 mutations in Drosophila are hypothesized to destabilize loop 3 and interfere with its ability to bind Ca2+. The mechanism by which these AD3 mutations induce loss of function in synaptotagmin can be elucidated through the use of computational methods and accelerated molecular dynamics.
Computational methods, and accelerated molecular dynamics simulations in particular, are able to model the behavior of proteins with less computational cost than unaccelerated dynamics. Accelerated dynamics reduce the simulation time by sampling the proteins configuration space more broadly. The accelerated model removes barriers in the protein's energy landscape by reducing the depth of local energy minima below a certain threshold, resulting in enhanced sampling. In this study, we have used accelerated molecular dynamics to simulate the effect of the mutations 180Y to 180F and 180Y to 180N on the stability of synaptotagmin C2A on the timescale of 100 nanoseconds, and also simulated the wild type (PDB 4wee) for comparison.
Based on our analysis of the hydrogen bonding across different acceleration parameters, we decided to use alpha=200 and dual=off in our final runs. We used our lab's crystal structure of synaptotagmin C2A and C2B to generate the mutants of interest with pymol's mutagenesis wizard. Each of the 6 resulting structures was run under different random seeds for 50,000,000 frames with a timestep of 2 femtoseconds (total of 100 ns) on lonestar 5. The pdb structures were solvated and ionized in a cube and minimized for 1000 frames. A single atom in the center of the protein was restrained to prevent drifting. We restrained residue 226 in C2A and 359 in C2B.
The alignment of synaptotagmin-1 through synaptotagmin-17 was computed with mafft. A python script cut.py was used to cut the alignment down to the region in which the AD3 mutation lies. The regions in the alignemt are location 352-357 in C2A and 510-516 in C2B. Mutations in this sequence (SDPYVK) are fatal and therefore it is highly conserved across all C2 domains. This indicates that the 6 residue sequence plays an important structural role in the protein and its ablility to interact with calcium.
Animations of the volume analysis can be found here